Energy Policy 67 (2014) 104–115

Contents lists available at ScienceDirect

Energy Policy

0301-42http://d

n CorrE-m

portugagtroncobernard

1 Te2 Te

journal homepage: www.elsevier.com/locate/enpol

Japan's energy conundrum: Post-Fukushima scenariosfrom a life cycle perspective

Joana Portugal Pereira a,d,n, Giancarlos Troncoso Parady b,1, Bernardo Castro Dominguez c,2

a United Nations University, Institute of Advanced Studies, 6F International Organizations Centre, Pacifico-Yokohama, 1-1-1 Minato-Mirai, Nishi-ku,Yokohama 220-8502, Japanb The University of Tokyo, Graduate School of Engineering, Department of Urban Engineering, 7-3-1 Hongo, Bunkyo-Ku, Tokyo 113-8656, Japanc Bernardo Castro Dominguez, The University of Tokyo, Graduate School of Engineering, Department of Chemical System Engineering, 7-3-1 Hongo,Bunkyo-Ku, Tokyo 113-8656, Japand Energy Planning Program, Graduate School of Engineering, Universidade Federal do Rio de Janeiro, Centro de Tecnologia, Bloco C, Sala 211, CidadeUniversitária, Ilha do Fundão, Rio de Janeiro, RJ 21941-972, Brazil

H I G H L I G H T S

� We developed a LCA integrated scenario approach to estimate environmental and economic co-benefits of energy mixes in Japan.

� Renewable rise and fossil fuel dependence reduction leads to LCA emissions decrease up to 34%.� A balance between renewables and nuclear power is a desirable alternative, from an economic, environmental and security of supply perspective.� Levelized costs of renewables are competitive with nuclear power.

a r t i c l e i n f o

Article history:Received 3 December 2012Accepted 22 June 2013Available online 27 July 2013

Keywords:Japanese energy portfolioEconomic-environmental co-benefitsLife cycle assessment

15/$ - see front matter & 2013 Elsevier Ltd. Ax.doi.org/10.1016/j.enpol.2013.06.131

esponding author. Tel.: +81 452 212 300.ail addresses: joanaportugal@gmail.com,l.pereira@ias.unu.edu, portugal.pereira@ppe.uso@ut.t.u-tokyo.ac.jp (G. Troncoso Parady),o@chemsys.t.u-tokyo.ac.jp (B. Castro Domingul.: +81 358 416 234.l.: +81 358 417 300.

a b s t r a c t

This study aimed at evaluating the co-benefit implications of alternative electricity generation scenariosin Japan, in a post-Fukushima context. Four scenarios were designed assuming different shares of energysources in a 2030 timeframe. Applying a life cycle assessment (LCA) methodology, scenarios wereassessed in terms of cumulative non-renewable energy (NRE) consumption, global warming potential(GWP), terrestrial acidification potential (TAP), and particulate matter formation (PMF). Additionallyelectricity generation costs were evaluated. Results demonstrate that the current dependence on fossilfuel is unfeasible in the long run, as it results in 14% higher NRE consumption, an increase of 32% on GHGemissions, 29% on TAP and 34% on PMF, and 9% higher cost than the baseline scenario under pre-Fukushima conditions. On the other hand, a share of up to 27% of renewable energies is technicallypossible and would result in a 34% reduction of NRE consumption, 29% decrease of GHG emissions, andcontribute to the mitigation of 24% of TAP and PMF impacts, at minor increase of levelized costs.Increasing the share of renewables and phasing-out thermal power would therefore increase theresilience of the Japanese economy toward external oil markets, cope with environmental protectionpriorities, while promoting economic development.

& 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Energy security of supply is on the top of the political agenda inJapan. Being the 3rd largest economy in the World, Japan exhibits an

ll rights reserved.

frj.br (J. Portugal Pereira),

ez).

intensive energy consumption pattern in order to keep its economyrunning. However, the country presents the lowest energy self-sufficiency ratio (4%) among industrialized nations, as it has virtuallyno endogenous fossil fuel resources and an incipient renewablemarket (Onoue et al., 2012). In 2011, the imports of crude oil werenearly 207 million kL, largely from Middle-East countries (87%) (IEEJ,2012), which led Japan to the podium of net oil importers.

The electricity generation sector accounts for one quarter oftotal final energy consumption. In 2010, the electricity supply bymajor producer companies totalized 906 TW h, up 6.29 times from1965 (FEPC, 2012a). Although the government virtually started theliberalization of the sector in 2000, the electricity market is still

Table 1Changes in nuclear dependence, increased cost of fossil fuels and net revenue of power companies in Japan before and post-3/11 (2010–2012 calendar year).

Power companies Share of nuclear power in the total power generation Increased cost of fossil fuels (million US$) Net revenue (million US$)

2010 (%) 2 011 (%) 2011 2012 (estimated) 2011 2012 (estimated)

Hokkaido 44 26 606 1,819 �873 �1394Tohoku 26 2 3,032 3,032 �2813 �2110Tokyo 28 10 10,671 12,490 �9483 �1880Chubu 13 2 3,032 2,668 �1116 �788Hokuriku 28 1 970 1,334 �61 �388Kansai 44 20 5,093 8,488 �2934 �7057Chugoku 3 9 0 970 24 �728Shikoku 43 19 849 2,425 �109 �1686Kyushu 39 16 3,032 5,699 �2013 �5445

J. Portugal Pereira et al. / Energy Policy 67 (2014) 104–115 105

controlled by 10 regional monopolies, which supply 85% of thetotal demand (FEPC, 2011). As established under the ElectricityBusiness Act (METI, 2005), these regional monopolies verticallyintegrate the generation, distribution and transmission of electri-city, which makes the penetration of independent producers ofrenewable energies challenging. As a result, the electricity gen-eration portfolio has been dominated by fossil fuel thermal andnuclear power, while renewables (mainly hydropower) have aminor contribution.

The triple disaster that hit Japan in March 11, 2011 (hereafterreferred to as 3/11) has threatened an already vulnerable electricsector. In the aftermath of the quake, nearly 30% of the supply wasput out of service. Although most units were back in service shortly,all of the 54 nuclear reactors, which account for a total of 47 GW,were gradually shut down for safety inspections or maintenance,yet as a result of public pressure, whether or not these units can beput back online is uncertain in the short-term. To avoid potentialblackouts and overcome this drastic cut of nuclear power, the shareof fossil fuel thermal was increased. As a result of this decrease inthe share of nuclear power and the consequent hike of 25.2% infossil fuel imports, all power companies but Chugoku (with lowreliance on nuclear power) reported net losses in 2011 and 2012(Table 1; Asahi Shimbun, 2012; FEPC, 2012b; Supply and DemandVerification Committee, 2012). At the national level, the conse-quences of this growing dependence on energy imports threateneconomic stability and environmental sustainability. In fact, afterthe accident, Japan logged a record 2.5 trillion JPY ($US 31 billion)trade deficit, the first since the 1979 oil shock, and largely as a resultof a jump in fossil fuel imports, putting further strains on an alreadyfragile economy (Vivoda, 2012).

Since the first oil shock in 1973 and until 3/11, nuclear powerhas been a key pillar to reinforce Japan's energy resilience againstoil market fluctuations and to mitigate climate change. In 2010, the“Strategic Energy Plan” defined the target of doubling the percen-tage of electricity generated by renewable sources and nuclear,and a 30% of reduction of energy-related CO2 emissions comparedto 1990 level by 2030 (METI, 2010; Duffield and Woodall, 2011).Under this strategy, the share of nuclear power in the electricitygeneration portfolio would ramp up to nearly 50% and 14 newreactors were planned to be built by 2030. This strategy wasoriginally scheduled for revision in 2013, but in light of 3/11disaster, last September 2012, the Energy and Environment Coun-cil (EEC) of Japan drew up the “Innovative Strategy for Energy andEnvironment” endorsing a gradual phase-out of nuclear power bythe 2030s (EEC, 2012a). Nevertheless, under pressure from thebusiness lobby, this strategy failed to get the cabinet's approvalhence falling short of becoming a legally binding document.Furthermore, the proposal was heavily criticized (see e.g. DeWit,2012; Kingston, 2012), as it did not reveal a concrete roadmap oran action plan to achieve this goal. It rather expressed an

ambiguous vision for a “society not dependent on nuclear powerin the earliest possible future”, a “realization of a green energyrevolution” and a “stable supply of energy”.

Japan needs to move towards a long-term reliable and afford-able energy strategy that guarantees security of supply, powerseconomic growth needs, supports climate change mitigationpolicies, and ensures the public's trust; yet the roadmap to reachthis goal is uncertain. Currently two antagonistic views are indebate regarding whether or not Japan should phase-out nuclearpower and which alternative sources should replace it; on the onehand, the central government, pressured by industrial lobbyists, apart of the so called ‘nuclear village’, supports the restart of thenuclear reactors, claiming that nuclear power is the only option tocope with economic needs at an affordable cost, especially givenshort-term supply constraints. On the other hand, local govern-ments and public opinion are committed to increase the share ofrenewable energies; they see the current challenge as an oppor-tunity to move towards a green economy, which would bringclimate resilience, regional development, green jobs, and reducethe energy dependence on imports and power monopolies. How-ever, renewable energies are still costly and its technologicalmaturity is yet to be fully developed. Japan is therefore facing adilemma. Can the government reach a turning point and switch toa green and resilient economy based on endogenous and renew-able energies? If so, what is the optimal electricity generation mixsupply portfolio so that economic and environmental co-benefitsare maximized?

Life cycle assessment (LCA) integrated with scenario design is arobust approach to determine the optimal energy portfolio, as itintegrates a wide range of electricity generation systems andevaluates its impacts under holistic lens. While pertinent to thefield, previous studies that evaluate electricity scenarios in Japanapplied a supply-demand analysis to determine the reliability andfeasibility of energy mixes, but they lacked in forecasting the fulllocal and global life cycle environmental impacts and economicimplications of energy mixes in Japan from a holistic view point(see Nakata, 2002; Takase and Suzuki, 2010; Onoue et al., 2012;Zhang et al., 2012). To the authors' knowledge, only Hondo (2005)and Bhattacharya et al. (2012) attempted to address these issues.While Hondo (2005) conducted a GHG–LCA that quantifies theglobal warming potential (GWP) of power generation systems inJapan, the cost implications of each system were not evaluated, oran optimal energy mix scenario was discussed. The cost implica-tions of energy mix scenarios were evaluated by Bhattacharya andhis colleagues through forecasting and backcasting methods. Theglobal environmental impacts and costs of several energy mixeswere assessed; nevertheless the study does not evaluate localenvironmental impacts.

To fill this gap, four prospective scenarios of alternativeelectricity generation mixes in a 2030 timeframe are developed

J. Portugal Pereira et al. / Energy Policy 67 (2014) 104–115106

with the objective of evaluating its implications in terms of globaland local environmental impacts, as well as generation costs;furthermore, the economic and social implications of these sce-narios are discussed in order to explore in a comprehensivemanner potential directions for the future of Japanese energypolicy in the aftermath of Fukushima Dai-Ichi accident. In thatsense, the following indicators have been evaluated: cumulativenon-renewable energy (NRE) consumption, GWP, terrestrial acid-ification potential (TAP), particulate matter formation (PMF), andgeneration costs.

2. Prospective scenarios for a future electricity generation mixin Japan

Departing from a pre-Fukushima state of affairs, four energymix scenarios were developed to evaluate alternative policydirections. Scenarios were constructed taking into consideration(i) the ongoing debates on the future direction of Japan's energypolicy, (ii) the current socio-economic and socio-political environ-ment of contemporary Japan, (iii) estimated potential for renew-able energy sources and (iv) technical constraints for theexpansion of renewable energy.

The main assumptions for the development of the scenariosconcern the distribution of nuclear energy and renewables, whichlie at the center of the ongoing discussion about Japan's energyfuture. These assumptions are largely based on assumptions madeby the Cabinet's Energy and Environment Council, which draftedthree alternative energy scenarios and presented them to thepublic in the form of public hearings at the end of June 2012 (seeEEC, 2012b).

Table 2Estimated potential from renewable sources and installed capacity of high renewable sSource: developed by authors based on ANRE (2012a).

Energy source Estimated potential (GW) Est

Biomass –

Solar-PV (Households, public facilities, factories) 8600Wind power (Onshore) 1500Geothermal (Including national parks) 4.3 (6.4) 26

Table 3Japanese energy mix scenarios in 2030 (generated electricity).

Energy source Pre-Fukushima Zero nuclear, highthermal

Zero nuclear,renewables

Baseline scenarioa

(%)Scenario 1 (%) Scenario 2 (%

Nuclear power 26.4 0.0 0.0Coal 24.3 34.0 27.8LNG 29.4 45.0 33.5HFO 10.2 11.3 11.7

Total fossil fuels 63.9 90.3 73.0Hydro (Run-of-the-river)

8.0 8.0 10.0

Hydro (Pumped-storage)

0.5 0.5 0.5

Biomass 0.3 0.3 2.1Solar-PV 0.2 0.2 5.6Wind power 0.4 0.4 6.0Geothermal power 0.3 0.3 2.8

Total renewables 9.7 9.7 27.0Total 100.0 100.0 100.0

a Baseline scenario sources: EEC (2012b) and ANRE(2012a).

Regarding nuclear power, in 2010, the baseline year, itaccounted for approximately 26% of total generated electricity.Against this background, two possible alternatives are put forth: azero nuclear option where all reactors are shut down and decom-missioned, and a nuclear reduction option which brings thegeneration percentage down to 15%. Both alternatives reflect to alarge extent the ongoing debate on nuclear energy policy, whereon the one hand, 70% of the country believes that its reliance onnuclear power should be reduced (PRC, 2012), while on the other,business and industrial groups such as the Japan Business Federa-tion (Keidanren) are urging the government to restart the reactorsas soon as possible citing the economic impact of energy shortagesand high energy prices on the industrial apparatus of the resource-poor nation.

An increasing share of nuclear power in the electricity genera-tion portfolio was not considered in the present study, as it seemsunlikely that all reactors will be put back online given stricterstandards for operation established by the Nuclear RegulatoryAuthority (NRA), and the re-evaluation of existing active fault linesaround plant sites. Furthermore, it seems improbable that givenattitudes from local governments and the general public, thegovernment will be able to build consensus for constructing newreactors.

In the nuclear reduction scenarios, spent fuel is assumed to berecycled, in line with current industry standards. High level wasteis vitrified and stored for 30–50 years before final disposal, andlow level waste is contained in metal drums and stored in a lowlevel radioactive waste storage center (FEPC, 2013).

Regarding renewable energy, in 2010, approximately 10% of thetotal generated electricity came from renewables, out of whichhydropower accounted for 8.50%, while other sources added up to

cenarios.

imated potential (GW h) Installed capacity high renewable scenarios(GW)

28,100 2.8893,000 51.2770,000 32.96

,000 (42,000) 3.85

high Nuclear reduction, lowrenewables

Nuclear reduction, highrenewables

) Scenario 3 (%) Scenario 4 (%)

15.0 15.028.6 22.034.7 26.712.0 9.375.3 58.08.0 10.0

0.5 0.5

0.3 2.10.2 5.60.4 6.00.3 2.89.7 27.0100.0 100.0

Fig. 1. Analytical framework and system boundaries of the study.

J. Portugal Pereira et al. / Energy Policy 67 (2014) 104–115 107

1.2% of the total. Against this state of affairs, two policy paths areput forth: a no expansion alternative where the share of renew-ables stays identical to 2010 levels, and a “High Renewable”alternative. An upper limit of renewable energy penetration wasassumed at 27%, a value that lies between the 35% maximumthreshold estimated by the Energy and Environment Council (EEC,2012b) and the more conservative estimate of 21% from METI's2010 Strategic Energy Plan (ANRE, 2012a). Previous research byZhang et al. (2012) had also suggested feasible market penetrationof renewables up to 30%, if thermal power systems are used asbase load to absorb supply and demand fluctuations. A similartrend was observed by Esteban et al. (2012a), revealing that 40% ofelectricity demand can be met by solar-PV and win power systems,during peak hours, if 40 TW of storage capacity is installed tobalance peak demand in the summer period in Japan.

The first scenario, “Zero Nuclear, High Thermal” considers asetting where all nuclear power plants are offline and their shareis compensated for with an increase in thermal power, particularlyheavy fuel oil (HFO) and liquid natural gas (LNG). These shares arebased on the Institute of Energy Economics of Japan's estimationsabout increase in fossil fuel procurement in 2012 (IEEJ, 2011), andreflect to a large extent the state of affairs of the Japanese energymix until July 1st, 2012 when the Ohi reactors were restarted.

The second scenario, “Zero Nuclear, High Renewables” assumesthat the nuclear share is compensated instead with an increase inrenewables while a decrease in fossil fuel thermal power. Based onestimated potentials for each energy source by the Agency forNatural Resources and Energy (ANRE, 2012a), wind power share isestimated at 6% of generated output, followed by solar power with5.6%, geo-thermal power with 2.8% and biomass with 2.10%.Table 2 presents estimated potential from renewable resourcesand installed capacity in the “High Renewable” scenarios.

The remaining two scenarios assume a reduction in the nuclearenergy share to 15%. Scenario 3, “Nuclear Reduction, Low Renew-ables” assumes no change in the renewable energy share, and anapproximate increase of 12% in the fossil fuel share to compensatefor the reduction in nuclear power. On the other hand, scenario 4,“Nuclear Reduction, High Renewables” assumes the same renew-able energy distribution as scenario 2, translating in a reduction ofalmost 6% in the share of fossil fuels. Scenario characteristics aresummarized in Table 3.

3 GWP is calculated as carbon dioxide equivalent (CO2e) by the followingexpression:CO2e¼CO2+23 �CH4+296 �N2O, according to the GWP factors suggestedby IPCC in its forth assessment report (Solomon et al., 2007).

4 TAP is calculated as sulfur dioxide equivalent (SO2e) by the followingexpression: SO3e¼SO2+2.45 �NH3+0.56 �NOx, as suggested by EcoInvent in itsReCiPe midpoint method (Goedkoop et al., 2008).

3. Methodology

3.1. Scope of the study

This study is comprised of two analyses: a life cycle assess-ment (LCA) of the environmental co-benefits and an estimation

of levelized generation costs of different electricity generationmix scenarios in Japan in a 2030 timeframe. The LCA, asdescribed in ISO 14040-44 guidelines (ISO, 2006), comprisesthe evaluation of the NRE consumption, GWP, TAP and PMF ofthe entire life cycle of electricity generation systems, referred toas, ‘Well-to-Meter’ (WTM) analysis. System boundaries includeboth upstream (extraction of fuels and raw materials, fuelprocessing and transportation) and downstream processes(operation of power plants to generate electricity and its trans-mission and distribution to end users) (Fig. 1). Additionally, itincludes the “Cradle-to-Gate“ cycle of thermal power plantinfrastructure construction, as well as the manufacture of mate-rial requirements for the construction of renewable powergeneration facilities, namely solar-photovoltaic (PV). panels andwindmills. The evaluation of the full life cycle of electricitygeneration is particularly important when evaluating fossil fuelsystems, as upstream processes can be as much as 25% of directenvironmental impacts of the power plant operation (Weisser,2007). Furthermore, the ‘Cradle-to-Gate’ cycle of power plantinfrastructure is a relevant stage in solar-PV, wind and hydro-power technologies, as the construction of facilities and theirsupporting processes are both resource and energy intensive(Varun et al., 2009).

Table 4 displays the electricity generation technologiesconsidered in this study. It includes nuclear pressurized waterreactor (PWR), coal steam turbine (ST), liquid natural gascombined cycle (CC), heavy fuel oil boiler, biomass co-generation boiler, run-of-the-river hydropower (RORH),pumped-storage hydropower (SPH), geothermal, wind powerand solar-PV. Energy systems were simulated in the globalemission model of integrated systems (GEMIS) software (ÖkoInstitute, 2012). GEMIS is an LCA software that allows users tomodel input/output streams of all stages of the life cycle(materials used, fuel extraction, processing and delivery). Data-base were customized, in terms of electricity generation tech-nology selection and net energy efficiency of processes, to makesure Japanese site-specific and 2030 timeframe data wereincorporated in the model.

The Functional Unit (FU) applied in the analysis is 1 kW he

of electricity supplied to end users. The co-benefits of scenarioswere evaluated in terms of NRE consumption, as primaryfossil energy consumed (ktoe/kW he), CO2e

3, SO2e4, and

Table

4Characteristicsof

electricityge

nerationsystem

sev

aluated

inthis

study.

Tech

nology

System

description

Plantnominal

capac

ity

(MW

e./y

r)a

Capac

ityfactor

(%)a

Plantlifetime

(yr)

b)

Net

energy

efficien

cy(η

e)(%

)b,c

Nuclea

rPW

RPW

Rwithan

enrich

men

tof

3.4%

andabu

rn-upof

40GW

D/t-U

.Nuclea

rfuel

recycled

.Highleve

lwaste

vitrified

andstored

for30

–50

yrs;

low

leve

lwaste

contained

indrumsan

dstored

instorag

ecenters

1250

7040

100

Coa

l-ST

Coa

l-ST

plantwithselectivecatalyticreduction(SCR),fluega

sdesulphurization

(FGD),op

eratingat

supercritical

conditions.

NoCCS

750

8040

45

LNG-CC

LNG-CCpow

erplantwithSC

Ran

dlow

NOxbu

rner

1350

8040

60HFO

-ST

HFO

-STpow

erplantwithSC

Ran

dFG

D45

050

4046

RORH

RORH

plantwithsm

allreservoir

1245

6010

0PS

HPS

Hplant,pumped

from

alower

reservoirto

ahigher

elev

ationforload

balancing

1245

6010

0Biomassco

-gen

eration

Com

bined

hea

tan

dpow

erplantusingwoo

dch

ipsas

feed

stoc

k;includes

fuel

storag

e10

8040

39So

lar-PV

RooftoptypePV

(3kW

)withaluminum

fram

ean

drack

.ThePV

cells

use

solar-grad

e(SOG)polycrystallin

esilic

on1.2

1235

100

Windpow

er(O

nsh

ore)

Windturbines

inaparkwith10

units(4.4

MW

each

);ex

cludes

cables,transformers

2020

2010

0Geo

thermal

Dou

bleflashtypege

othermal

stea

mturbine.

Includes

well(lessthan

1000m

dee

p)dev

elop

men

tan

ddrillingfailu

re30

8040

100

aEE

C(201

2b).

bGóm

ezan

dW

atterson

(2006

).cIn

this

study,

net

energy

efficien

cyis

defi

ned

asthepercentage

offossilfuel

inputen

ergy

that

isretrieve

das

electricityou

tput.Th

erefore,

renew

able

energy

system

spresentavirtual

net

energy

efficien

cyof

100%

.

J. Portugal Pereira et al. / Energy Policy 67 (2014) 104–115108

PM10e5emissions released (gpollutant/kW he) and generation costs

(US$2012/MW h) of the electricity WTM life cycle. Additionally,impacts were evaluated in terms of total electricity demand in2030 (Section 3.2).

3.2. Electricity demand forecast

The electricity demand in Japan has risen drastically, associatedwith the fast industrial growth over the past decades. Frommid-1990s, however, consumption has stabilized given the eco-nomic slowdown and population decline. In the future, demand ispredicted to remain fairly steady, as population is projected tocontinue declining and the economic panorama is yet uncertain.

Demand of electricity depends upon a wide range of factors,including the socio-economic conjecture of a country, technologi-cal development, climate conditions, consumer behavior, andenergy policy framework. Generally, models to predict energydemand are classified as top-down and bottom-up methodologies.Top-down methods tend to adopt a more generic approach thatfocuses on an aggregated level of analysis. On the other hand,bottom-up approaches take into consideration detailed activitydata at end-use levels for which demand is forecasted.

In this study, a top-down approach has been adopted; percapita electricity demand in 2030 was estimated by fitting amultiple linear regression on historical data from 1980 to 2010,using GDP per capita, and industrial and residential price ofelectricity as predictors of per capita demand. While energyconservation policies play a major role in regulating energydemand (clearly visible in the aggressive energy saving campaigns(setsuden) in post-3/11), this parameter was not taken into con-sideration in the present study, as future policy are yet to bedefined, which would increase uncertainty in the model. The mo-del is, therefore, specified as: yt¼β0+β1 �GDPt+β2 � indpricet+β3 �respricet+εt, where yt is the electricity demand per capita in yeart, GDP is the gross domestic product per capita in year t in 2000PPP dollars, and indprice and resprice are the electricity tariffs forindustries and households in year t, respectively. Historical GDPper capita data has been obtained from the World DevelopmentIndicators database (WB, 2013), whereas forecasts were obtainedfrom JCER (2013), which estimates a GDP growth rate in Japan of0.4 and 0.2, between 2011 and 2015, and 2016 and 2025,respectively. Historical residential and industrial tariffs in Japanhave been gathered by IEA (2010), and estimations assume agradual increase of prices by 20% until 2030, given the foreseenincrease in share of renewable and reduction of nuclear energy inthe generation mix (JCER, 2013).

The overall aggregated electricity demand in 2030 was obtainas the product of estimated electricity demand per capita andpredicted population. Demographic historical statistics and fore-casts were obtained by the National Institute of Population andSocial Security Research (IPSS, 2012; 2013). Table 5 summarizesthe regression estimation results.

Adjusted R2 of the regression is 0.97, suggesting a very goodpredictive power. Overall, aggregated electricity demand in Japanin 2030 is estimated at 916.68 TW h, 9% lower than in 2010 levels.This is in range with the National Policy Unit (2012) projections,which estimates a total electricity consumption of 1000 TW hby 2030.

5 PMF is calculated as particulate matter o10 μm (PM10e) by the followingexpression:PM10e¼PM10+0.2 � SO2 +0.32 �NH3+0.32 �NOx, according to the factorssuggested by EcoInvent in its ReCiPe midpoint method (Goedkoop et al., 2008).

Table 6Cumulative NRE consumption of electricity generation systems (toe/kW he).Source: developed by the authors based on GEMIS databases (Öko Institute, 2011).

Lifecycle stage Nuclear Coal LNG HFO RORH SPH Biomass Solar-PV Wind Geo-thermal

Upstream 0.094 0.091 0.117 0.103 0.000 0.000 0.004 0.000 0.000 0.000Downstream 0.004 0.115 0.065 0.110 0.000 0.000 0.000 0.000 0.000 0.000Infrastructure 0.000 0.000 0.000 0.000 0.007 0.006 0.000 0.004 0.004 0.000

Total 0.098 0.206 0.182 0.214 0.007 0.006 0.004 0.004 0.004 0.000

Table 7Global warming potential of electricity generation systems (gCO2e/kW he).Source: developed by the authors based on GEMIS databases (Öko Institute, 2011).

Lifecycle stage Nuclear Coal LNG HFO RORH SPH Biomass Solar-PV Wind Geo-thermal

Upstream 13.80 49.10 112.71 57.66 0.00 0.00 11.48 0.00 0.00 0.00Downstream 0.00 798.86 333.85 625.83 0.00 0.00 3.11 0.00 0.00 10.19Infrastructure 3.15 2.23 0.37 1.96 58.41 54.76 3.52 60.92 25.04 2.88

Total 16.95 850.18 446.93 685.46 58.41 54.76 18.11 60.92 25.04 13.07

Table 8Terrestrial acidification potential emissions of electricity generation systems (gSO2e/kW he).Source: developed by the authors based on GEMIS databases (Öko Institute, 2011).

Lifecycle stage Nuclear Coal LNG HFO RORH SPH Biomass Solar-PV Wind Geo-thermal

Upstream 0.13 0.13 0.52 0.57 0.00 0.00 0.02 0.00 0.00 0.00Downstream 0.00 0.90 0.85 1.10 0.00 0.00 0.94 0.00 0.00 2.84Infrastructure 0.01 0.00 0.00 0.00 0.09 0.08 0.01 0.07 0.05 0.01

Total 0.13 1.03 1.38 1.67 0.09 0.08 0.97 0.07 0.05 2.84

Table 5Per capita electricity demand regression estimation results.

R2 0.974Adjusted R2 0.971SSE 0.190N 30Variable name β S.E. t stat p value 2030 input valuesConstant 2.290 1.311 1.747 0.09 –

GDP (Thousand PPP 2000$) 0.153 0.022 6.828 0.00 41.44Industrial Electricity tariffs (JPY/kW h) -0.438 0.190 -2.312 0.03 16.26Residential Electricity tariffs (JPY/kW h) 0.260 0.140 1.859 0.07 24.44

Significance level: n10%, nn5%, nnn1%.

J. Portugal Pereira et al. / Energy Policy 67 (2014) 104–115 109

3.3. Environmental inventory analysis

As stated above, the WTM analysis includes three different stages:(i) upstream processes, (ii) operation of power plants, and (iii) cradle-to-gate infrastructure life cycle. Therefore, co-benefits were calcu-lated as the sum of all corresponding sub-stages of each electricitygeneration system, as follows: IWTM¼∑fe � (Iu,e+Id,e+Ii,e), whereas ferefers to the share of each energy source e in the designed scenariosas defined in Table 2, Iu,e reflects the impacts associated to upstreamprocesses, Id,e accounts for the impacts during the operation of powerplants to generate electricity, and Ii,e includes the impacts of thecradle-to-gate infrastructure life cycle. Following sections detailcalculations and data sources of each stage.

3.3.1. Upstream stageNRE consumption and air emissions of energy resources

extraction and processing were estimated by the following expres-sion Iu,e¼Ae/ηe, where Ae refers to the energy used and environ-mental burdens to extract and process one unit of fossil fuels

and ηe refers to the net thermal efficiency of power plants, asstated in Table 3.

3.3.2. Downstream stageThe downstream stage refers to the impacts associated to the

power plant operation. NRE consumption is given as the ratio ofthe process efficiency and the energy losses in Transmission andDistribution (T&D) systems: NRE¼1/(∑fe/ηe)� (1� l), where ferefers to the share of each energy source, ηe describes the powerplant net thermal efficiency and l the losses of T&D.

The estimation of emissions of fossil fuel energy technologiesinvolves two different approaches, as suggested by Gómez andWatterson (2006) and EPA (2008). The assessment of CO2 and SO2

emissions follows a fuel input analysis, as emissions are mainlydependent on fuel carbon and sulfur content. Thus, emissions perkW he generated, transmitted and distributed are based on the LowHeating Value (LHV) of each fuel i, the carbon (Ci) and sulfur (Si)content of combusted fuel, oxidation fraction (OFi), the power plantnet thermal efficiency (ηe), and losses of transmission and distribu-tion (l): ECO2 ¼∑LHVi �Ci.OFi �1/ηe �44/12/(1� l), ESO2 ¼∑LHVi � Si.OFi �

Table 9Particulate matter formation potential of electricity generation systems (gPM10e/kW he).Source: developed by the authors based on GEMIS databases (Öko Institute, 2011).

Lifecycle stage Nuclear Coal LNG HFO RORH SPH Biomass Solar-PV Wind Geo-thermal

Upstream 0.04 0.07 0.24 0.16 0.00 0.00 0.01 0.00 0.00 0.00Downstream 0.00 0.33 0.49 0.35 0.00 0.00 0.43 0.00 0.00 0.57Infrastructure 0.00 0.00 0.00 0.00 0.07 0.07 0.00 0.05 0.03 0.00

Total 0.04 0.41 0.73 0.51 0.07 0.07 0.45 0.05 0.03 0.57

J. Portugal Pereira et al. / Energy Policy 67 (2014) 104–115110

1/ηe �64/32/(1� l). On the other hand, other pollutants are site specificand mainly depend on power plant technology, combustion char-acteristics, usage of pollution control equipment and environmentalconditions. Hence, they have been calculated on an Emission Factor(EFi) (Gómez and Watterson, 2006), as follows: Ep¼EFi/ηe � EF/(1� l).Air emissions of renewable energy technologies are negligibleduring the plant operation, except for biomass and geothermalpower plants. Although carbon dioxide emissions released duringthe operation of biomass power plants are not accounted for, asthey are admitted to be carbon neutral, significant amount of otherair pollutants, including N2O, PM10 and SO2, are emitted duringcombustion of wood chips (Cherubini, 2010). As for geothermalpower plants, CO2 and SOx emissions are released when diggingwells and maintaining hot water pipes (Hondo, 2005).

3.3.3. Infrastructure ‘cradle-to-gate’ life cycleThe ‘cradle-to-gate’ life cycle of power plant infrastructure was

estimated based on the ratio of the amount of raw materials needto construct each plant (mr), its associated fossil fuel energyconsumption and emission factors (EFr), and the total electricitysupplied yearly in each power plant (Q), as shown in the followingexpression: Ii,e¼∑(∑(mr � EFr)/Qe.

3.3.4. Cumulative NRE consumption and GHG emissions of evaluatedenergy systems

Tables 6–9 summarize the inventory analysis of the cumulativeNRE consumption and GHG, SO2e and PM10e emissions of theelectricity generation WTM life cycle, respectively. Results wereobtained by integrating adjusted GEMIS unitary process databasesinto electricity generation systems.

As expected, fossil fuel based electricity generation systemscontributes the most to the total environmental loads, rangingbetween 0.182–0.214 toe, 447–850 gCO2e, 1.03–1.67 gSO2e, and0.41–0.73 gPM10e/kW he of electricity supplied to end users. Thedownstream stage of fossil fuel based energy systems is the mostimpactful accounting for 62–94% of total environmental burdens,followed by upstream processes, while the infrastructure ‘cradle-to-gate’ life cycle plays a minor role. The nuclear based electricityreveals lower NRE consumption and GHG emissions than fossilfuel technologies and the lowest TAP and PMF potentials, beingtherefore competitive with renewable energy systems. Its impactsare mainly related to the upstream processes, for uranium extrac-tion and enrichment. However, if human toxicity and radioactivepotential impacts related to waste disposal and plant decommis-sion were accounted in the present analysis, nuclear technologymigth have a higher impact (Valentine, 2011).

Renewable energies, on the other hand, reveal lower NREconsumption and GHG emissions than fossil fuel systems, beinggeothermal and biomass technologies the most attractive. NREconsumption ranges between minimal values for geothermalsystems to 0.007 toe/kW he for hydropower; whereas GHG emis-sions vary between 13.07 and 60.92 gCO2e/kW he. In terms of localenvironmental impacts, geothermal and biomass systems presentloads at the same range than fossil fuel thermal technologies and

higher than nuclear power. These derive from the pipeline emis-sions released during downstream stages, mainly PM10, SOx andNOx during combustion of biomass and SOx from digging wells andmaintaining hot water pipes in geothermal plants. On the otherhand, solar-PV, wind power, and hydro systems are the mostattractive ones among renewables with regards to local environ-mental impacts.

3.4. Costs estimation

The cost estimation methodology consisted on using a startingpoint price for the overnight capital, fixed and variable costs(Tidball et al., 2010). These base costs were scaled to the pre-specified nominal plant capacity using rationing by capacity withthe assumption of Williams' six-tenths rule, which assumes thatthe relationship between costs and size of a manufacturing facilityincreases by an exponent of six-tenths, and for the capital costs alinear exponent on the operating costs (Williams, 1947). Thesemethods have provided reliable and accurate cost projectionsin chemical plants, and can be extrapolated to power plants.Moreover, the producer's price index (PPI) was used to adjustthe time value of the costs. In order to ensure that the scaled costswere reasonable, different datasets contained in Tidball et al.(2010) and EIA (2010) were used to compare and adjust theestimates. An overestimated cost was substituted by the immedi-ate value of the upper limit found in the literature; while ifunderestimated, the lower limit was used. Levelized costs (LCOE)were calculated using the simplified formula suggested by Shortet al. (1995):

sLCOE¼{(overnight capital cost� capital recovery factor+fixedO&M cost)/(8.76� capacity factor)}+variable costs f (O&M cost+fuel cost+CO2 costs).

The capital recovery assumed an interest rate of 3% and thecapital recovery factor was determined from the statutory year.These values, as well as the upstream stage of raw materialextraction and processing fuel and CO2 emission costs wereobtained from the Cost Validation Council (2012) in order tocustomize for the Japanese context. Furthermore, since uncer-tainty is constantly present when estimating the costs of a newpower plant; to minimize this error and ensure reliability, theassessment was measured up to an evaluation performed by theCost Validation Council.

It is important to underline, however, that while environmentaldamage has been analyzed, its external costs are out of the scopeof this study. Therefore, costs associated with environmental andhuman health, and material damage from harmful inorganicpollutants, acidification precursors, and GHG have not been takeninto consideration. Furthermore, uncertainties related to indirecteffects of climate change, such as increasing frequency andintensity of tropical cyclones, as stated by Esteban et al. (2012b)has been not included in the present study.

Cost estimates, as presented in Table 10, exhibit a maximumdeviation of 30% from the Cost Validation Council estimates. Hydroand nuclear power plants present the lowest costs, at 92 and 93 US$/MW he, respectively; while, as expected, the oil-fueled power plant

Table 10Breakdown of cost estimates of electricity generation systems in 2030.

Tech. Units Nuclear Coal LNG HFO RORH SPHe) Bio-mass Solar-PV Wind Geo-thermal

Capital costs US$2012/kWb 4573 3214 1525 563 7522 – 3000 2500 2090 7360Variable US$2012/MW ha 1 3 7 8 4 – 7 0 0 0Fixed US$2012/kW 157 59 35 63 19 – 73 12 42 147Fuel costs US$2012/MW ha 8 49 90 186 0 – 155 0 0 0Statutory life Years 16 15 15 15 40 – 15 17 17 15CO2 costs US$2012/MW ha 0 38 15 29 0 – 0 0 0 0Total LCOF (2030) US$2012/MW ha 93 136 135 265 92 153 208 192 115 109Total LCOF (2030) JPY2012/kW ha 7.5 10.9 10.8 21.2 7.4 12.2 16.6 15.3 9.2 8.7Estimated costs for 2010Total LCOF (2010) US$2012/MW ha 129 147 135 235 108 153 231 524 159 155Total LCOF (2010) JPY2012/kW ha 10.4 11.8 11.4 18.9 8.7 12.2 18.5 42.0 12.7 12.4

a Costs for SPH were obtained from Matsuo (2012). It includes the pump storage generation system and the required nuclear electricity to pump the water back up.b 1 kW h is equivalent to 8.76 kW, assuming an operating time of 24 h throughout 365 days per year.

Fig. 2. Cumulative NRE consumption and global warming potential (GWP) of proposed scenarios (2030).

J. Portugal Pereira et al. / Energy Policy 67 (2014) 104–115 111

exhibit the highest costs (265 US$/MW he) as a direct result of highfuel prices (70% of total LCOE). High costs are followed by biomass(208 US$/MW he) and solar-PV power (192 US$/MW he), since thesetechnologies require high capital costs for higher nominal capacities.

Table 10 also shows that when compared to 2010 estimatedcosts, most technologies exhibit reduction in generation costs,largely as a result of learning curves of technological improve-ment. The largest differences are observed on renewable energysources (excluding hydro), particularly solar-PV, as a result ofsteeper learning curves given non-matured stages of development.Slight reductions in generation costs are also observed for fossilfuels, with the exception of HFO. Nevertheless it is important tohighlight that although an increase in fuel costs was forecasted toestimate costs, fossil fuel prices are highly volatile, and generationcosts are very sensitive to changes in fuel prices, hence theseestimates should be interpreted carefully. Regarding nuclearenergy generation costs, policy variables that might be reflectedon costs, such as the strengthening of countermeasures againstnatural disasters and decommission of Fukushima nuclear powerplant reactors are not included in the analysis.

4. Results

4.1. Global and local environmental impacts

This section discusses the life cycle impact assessment resultsof the baseline and alternative scenarios in terms of cumulative

NRE consumption, GHG emissions and generation costs of totalsupplied electricity in 2030.

Fig. 2 displays the total amount of fossil fuels consumed andGHG emitted to meet the electricity demand in 2030. If theJapanese Government returns to the pre-Fukushima electricitygeneration mix (baseline scenario), the total NRE consumption andGHG emissions would be 136 Mtoe and 352 MtCO2e, respectively.Under this scenario, the main source of fossil fuel consumptionand GHG emissions are thermal energy technologies (96%), whilenuclear and renewables play a minor role. Among alternativescenarios, scenario 1, “Zero Nuclear, High Thermal”, reveals thehighest NRE consumption (156 Mtoe) and GHG emissions(466 MtCO2e), a rise of 15% and 32%, respectively, when comparedto the baseline scenario. Similarly, scenario 3, “Nuclear Reduction,Low Renewables”, which replicates the case when nuclear reduc-tion is achieved by increasing the share of fossil fuels, denotes anincrease of fossil fuel dependence and GWP by 7% and 16%,respectively. Scenario 2, “Zero Nuclear, High Renewables”, on theother hand, which portrays a nuclear phase-out, higher penetra-tion of renewables, and reduction of fossil fuel dependence, resultsin a drop of 17% on NRE and 27% on GHG emissions. Scenario 4,“Nuclear Reduction, High Renewables”, exhibits the lowest fossilfuel dependency among all alternative scenarios, yielding consid-erable savings, and suggesting a mitigation potential of 29–34%compared with the baseline scenario.

Similar trends are followed in terms of local environmentalimpacts, measured as TAP and PMF potential (Figs. 3 and 4). Thebaseline scenario shows a TAP of 0.8 MtSO2e and PMF potential of

Fig. 3. Terrestrial acidification potential potential (TAP) of proposed scenarios (2030).

Fig. 4. particular matter formation potential (PMFP) of proposed scenarios (2030).

Fig. 5. Levelized costs of proposed scenarios (2030).

J. Portugal Pereira et al. / Energy Policy 67 (2014) 104–115112

J. Portugal Pereira et al. / Energy Policy 67 (2014) 104–115 113

0.3 MtPM10e. The impacts are mainly due to LNG, coal and HFOthermal power systems, as they have a higher share in the energymix and present higher emission factors than nuclear and renew-able energy systems.

Scenario 1, “Zero Nuclear, High Thermal”, and scenario 3,“Nuclear Reduction, High Renewables”, displays a considerableincrease of SO2e and PM10e emissions when compared to thebaseline, due to greater dependency on fossil fuel thermal power.This suggests that even if nuclear power is reduced and substi-tuted by renewable energies, but fossil fuel thermal share in theenergy mix remains constant, the overall local environmentalimpacts of the electric sector will not change. In contrast, Scenario4, “Nuclear Reduction, High Renewables”, suggests a signi-ficant reduction of environmental impacts, as TAP and PMF fallby 24%. Thus, only if fossil fuel dependence and nuclear powershare are simultaneously reduced, environmental impacts will bemitigated.

4.2. Generation costs

This section describes the final distribution expenses based onthe LCOE results. Fig. 5 presents the breakdown of LCOEs of allscenarios as compared to the baseline scenario. Using a baselinecase at US$149 thousand million, scenario 2, “Zero Nuclear, HighRenewable” is the most expensive one, increasing by 9% up to US$163 thousand million, mainly as a result of the high capital costsrequired to implement these technologies.

Scenario 1, “Zero Nuclear, High Thermal” presented a LCOE ofUS$160 thousand million and it is slightly cheaper than scenario 2,“Zero Nuclear, High Renewables”. This indicates that renewablesmight be competitive in economic terms with fossil fuels by 2030;however, removing nuclear from the energy mix increases thegeneration costs. For instance, scenario 3 “Nuclear Reduction, LowRenewables” and scenario 4, “Nuclear Reduction, High Renew-ables” have estimated LCOEs of US$157 and US$154 thousandmillion, respectively. The reduction of nuclear energy representsincreases of 5% and 4% for the scenarios 3 and 4 with respect to thereference case.

When comparing scenario 3 “Nuclear Reduction, Low Renew-ables” and scenario 4 “Nuclear Reduction, High Renewables”,although the capital costs of renewables are significantly higherthan other energy sources, due to a reduction in the share of highthermal energy, scenario 4 turns out to be a cheaper option. This ismainly a result of high cost of fossil fuels, suggesting that by theyear 2030, a high renewables scenario could be competitive with ahigh nuclear scenario such as the baseline case. This makes itfeasible to provide a cost-effective solution to energy security ofsupply and climate change, without undermining economic devel-opment. It is important to note that project costs do not necessa-rily indicate the economic liability of a venture, but rather thecomparison to other alternatives. In this study the contrast ofdiverse scenarios helped to reveal the economic feasibility of apotential changes.

5. Discussion

In order to put in perspective the findings presented above, abrief overview of the current socio-political environment in Japansurrounding the drafting of Japan's new energy policy is firstdiscussed. Economically, the impact of the Fukushima accident isyet to be fully quantified, nevertheless, in terms of energy importsalone, in 2011, Japan's demand for LNG and HFO increased by 15.9%and 2.9%, respectively, when compared to 2010 (ANRE, 2012b) tocompensate for the shutdown of its nuclear reactors. This increasein imports is not only translated into an increase of generation

costs that was passed on to the final users in the form of a raise inthe energy bill by the Tokyo Electric Power Company (TEPCO), butalso it is identified as the main cause of Japan's first trade deficit inthe last 31 years (Vivoda, 2012). Consequently, the Japan BusinessFederation (Keidanren), Japan's largest business lobby, rallied for aquick restart of the idled reactors in order to reduce the adverseeffects of higher energy prices on business and industry sectors.This is a valid point taking into consideration that (i) Japan'seconomic growth has been driven by exports since the post-warperiod, and (ii) Japan's most powerful export industries, such asmachinery, transport equipment, and chemicals are also the mostenergy intensive ones. For example, chemical and steel industries,while accounting for 10.2% and 5.5% of Japan's total exports,respectively, add up to 65.3% of the total industrial energyconsumption (Schaede, 2012). Other potentially negative effectsof a growing dependence on imported fossil fuels are: (i) theincreased vulnerability to exogenous events, like the oil shocks ofthe 1970s, and (ii) a potential threat to Japan's national security interms of guaranteeing a stable supply of energy, an issue that washighlighted as a result of Iran's threat to close the Hormuz strait.

On the other hand, the public opinion regarding the use ofnuclear power has dramatically changed since 3/11. In 2007, only7% of the Japanese citizens supported a zero nuclear energy policy,while 21% favored a reduction; 53% supported maintaining currentlevels, and 13% agreed with an increase in the nuclear energyshare. A month after the accident, a change in the citizens' attitudetowards nuclear energy was already evident, with 11% demandinga total stop of reactors and 30% supporting a reduction (AsahiShimbun, 2011). The most recent public hearings conducted by thegovernment between July and August of 2012 to gauge the publicsentiment regarding nuclear energy revealed that among the threealternative scenarios proposed by the EEC, more than 70% of thepublic favored the zero percent nuclear scenario (Mainichi Daily,2012). The change in the public sentiment was also evident in theseries of anti-nuclear demonstrations that occurred all over Japansince 3/11 and continue until the time of writing of this article,constituting the largest public demonstrations since the 1960s.

This has led the government to announce a phase-out ofnuclear energy by 2030, only to take it back a day later, afterprotests from the Japan Business Federation (Keidanren). This isevidence of the existing tensions between economic and socialinterests and between different stakeholders involved in theenergy policy making process.

Based on this context, the analysis of the results presentedearlier suggests several implications:

(i)

A short-term transition towards a high fossil fuel thermalscenario and phase-out of nuclear, although desirable interms of the public sentiment towards nuclear energy, trans-lates into both higher economic and environmental costs.Additionally, a long-term dependence on fossil fuel seemsunrealistic given the increasing trend and volatility in fossilfuels prices, as earlier discussed. Furthermore, as the Bank ofJapan pursues an aggressive quantitative easing policy in a betto beat chronic deflation, at least in the short term, the JPY isforecasted to depreciate against the US$, which will consider-ably raise the costs of fossil fuel imports.

(ii)

The elimination of nuclear power from the electricity genera-tion mix (as shown in Scenarios 1 and 2) seems unfeasible ineconomic terms given its higher costs when compared to thebaseline scenario. A balance between renewables and nuclearpower seems to be the most desirable alternative, not onlyfrom an economic and environmental perspective, but alsofrom an energy security point of view. Social and technicalconsiderations also favor this balance; on the one hand, publicopinion and safety issues regarding nuclear reactor operation

J. Portugal Pereira et al. / Energy Policy 67 (2014) 104–115114

and its vulnerability to low frequency high impact events suchas 3/11 might hamper an aggressive nuclear policy such as theone advocated by the government in its “Strategic EnergyPlan”. On the other hand, renewables are still in an initial stateof development and there are significant constraints to beovercome regarding electricity storage, backup power anddistribution systems if a larger market penetration is to beachieved.

(iii)

Although cost wise nuclear reduction scenarios imply slight costescalation when compared to the baseline scenario, in the longrun, and from an environmental perspective, the “NuclearReduction, High Renewables” scenario, clearly presents largerenvironmental gains as a result of reduced NRE dependence,GHG and other local air pollutant emissions, when compared tothe “Nuclear Reduction, Low Renewables” scenario.

6. Final remarks and policy recommendations

This study aimed to evaluate the co-benefit implications ofalternative electricity generation scenarios in a post-Fukushimacontext. Four scenarios were designed assuming different sharesof energy sources in a 2030 timeframe. Applying a LCA methodol-ogy, these scenarios were assessed in terms of cumulative NREconsumption and GHG emissions. Additionally, levelized genera-tion costs were estimated.

In terms of policy making, a dynamic approach should considerthe energy problematic in both short-term and mid/long-termperspectives. Results revealed that the short-term strategy fol-lowed by the Japanese Government of idling nuclear power andsubstituting it with fossil fuel technologies (as modeled in scenario1) is not sustainable both in environmental and economic terms,as it results in an increase of 14% of cumulative NRE consumption,32% of GHG, 29% of TAP, 34% of PMF, and it is 8% more costly thanthe pre-Fukushima baseline scenario. Not only does this raisesconcerns related to Post-Kyoto Protocol commitments, but alsomakes the Japanese economy more vulnerable to unpredictablefluctuations of international fossil fuel prices and changes inexchange rates. On the other hand, due to infrastructure andtechnical constraints, an immediate transition to renewable ener-gies is not realistic in the short-term. In that sense, taking intoconsideration the public sentiment toward nuclear power and theeconomic and environmental implications of fossil fuel use, areduced share of nuclear energy is the optimal choice. However,given the public distrust in the nuclear regulatory institutions, theJapanese government should ramp up its efforts to guarantee thesafety of its nuclear reactors up to global standards of safety, publichealth and welfare, and reform the regulatory system as recom-mended by the Investigation Committee on the Accident atFukushima Nuclear Power Stations of Tokyo Electric Power Com-pany (Hatamura et al., 2012).

In the mid/long-term, a shift from non-renewable resourcedependence to endogenous energy sources might prove a success-ful strategy on environmental and economic terms. In that sense,the transition might be achieved from either nuclear powergeneration or renewables, as both energy systems are cost-effective alternatives to reduce foreign energy dependence, GHGemissions and other local air pollutants at an affordable cost. Asdemonstrated in scenario 4, “Nuclear Reduction, High Renew-ables”, an incremental share of renewables, together with areduction of nuclear power and fossil fuel thermal technologies,would result in 29% savings in NRE consumption and decrease of34% of GHG emissions, which would reinforce forthcoming Post-Kyoto Protocol commitments. Furthermore local environmentalimpacts, including TAP and PMF would decrease by 24%. Althoughwhen compared with the base scenario, generation costs areindeed higher, these costs might also may be understood as aninternalization of environmental externalities.

That being said, the large scale deployment of renewables alsofaces some constraints: (i) the initial necessary investments onrenewable energy infrastructure might impose economic limita-tions; (ii) regions with high renewable energy potential, particu-larly wind energy (Hokkaido and Tohoku areas), are rather farfrom high demand areas in Japan, such as the Tokaido megalopoliscorridor; and (iii) high shares of renewables might be limited tosome extent due to the fluctuating nature of generation, whichmight compromise a reliable steady supply of electricity, and req-uire significant storage capacity. This situation is also exacerbatedby the lack of connectivity with other regions or countries thatmight serve as a backup supply source. Given these constraints, itseems technically unfeasible to do without nuclear energy ifenvironmental targets are to be achieved; hence scenario 4,“Reduced Nuclear-High Renewables” is recommended as theoptimal alternative.

To accelerate this transition, it is desirable that the Japanesegovernment adjust its fiscal policy to develop market-basedinstruments that support the high capital costs of renewables,and conduct an institutional reform to accelerate the liberal-ization of the energy supply market thus breaking the virtualregional monopolies that control the energy markets in Japan. Inaddition, further investments on R&D are required to promotetechnological development and innovation in the renewableenergy sector.

Certainly, some policy measures are already being implemented,the latest example being the Feed-in-Tariff scheme which enteredinto effect on July 1st of 2012. Furthermore, the liberalization ofthe energy market is on the discussion table, particularly theseparation of the generation and the distribution systems. In thenear future, vital decisions will take place when designing the newenergy strategy for 2030. As demonstrated in this study, theGovernment of Japan has the potential and capacity to turn thecurrently energy crisis into an opportunity and move towards anenvironmentally friendly and economically competitive electricitygeneration mix.

References

ANRE, 2012a [Enerugii mikkusu no sentakushi no sakutei ni muketa saisei kanou enerugiikankei no kiso shiryou] [Fundamental Documents on Renewable Energy for the Establishment of Energy Mix Alternatives] (In Japanese). Agency for Natural Resourcesand Energy.

ANRE, 2012b. [Heisei 23 nendo enerugii juyou jisseki (soukoku)] [Energy Demand in Fiscal Year 2011 (News Release)] (in Japanese).Agency for Natural Resources and Energy.

Asahi Shimbun, 2011. [Genaptsu [Herasu yameru] 41% Asahi Shimbun seron chousa] [Asahi Newspaper Opinion Poll:41% Favor Reduction or Elimation of Nuclear Power] (in Japanese). Asahi Newspaper, 11 April. Available at: ⟨http://www.asahi.com/special/08003/TKY201104170324.html⟩ [accessed 10 September 2012].

Asahi Shimbun, 2012. Electricity supply sufficient even if next summer is a scorcher. Asahi Newspaper, 25 October. Available at: ⟨http://ajw.asahi.com/article/economy/business/AJ201210250073⟩ [accessed 25 October].

J. Portugal Pereira et al. / Energy Policy 67 (2014) 104–115 115

Bhattacharya, Janardhanan, Kuramochi, 2012. Balancing Japan's Energy and Climate Change Goals: Exploring Post-Fukushima Energy Supply Options. Institute of GlobalEnvironmental Studies, Japan.

Cherubini, 2010. GHG balances of bioenergy systems—Overview of key steps in the production chain and methodological concerns. Renewable Energy 35 (7), 1565–1573.Cost Validation Council, 2012. [Hatsuden cosuto shiito] [Energy Generation Cost Sheet] (in Japanese). National Policy Unit. Government of Japan.

Available at: ⟨http://www.npu.go.jp/policy/policy09/archive02_shisan_sheet.html⟩ [accessed 20 June 2012].DeWit, 2012. Japan's energy policy at a crossroads: a renewable energy future? The Asia-Pacific Journal 10 (38), 4.Duffield, Woodall, 2011. Japan's new basic energy plan. Energy Policy 39, 3741–3749.EIA, 2010. Updated Capital Cost Estimates for Electricity Generation Plants. U.S. Energy Information Administration.EEC, 2012a. Innovative Strategy for Energy and the Environment. Energy and Environment Council. National Policy Unit. Government of Japan.EEC, 2012b. [Enerugii kankyou ni kan suru sentakushi] [Alternatives on energy and the environment] (in Japanese). Energy and

Environment Council. National Policy Unit. Government of Japan.EPA, 2008. Direct Emissions from Stationary Combustion. Report no. EPA430-K-08-003. Environmental Protection Agency.Esteban, Zhang, Utama, 2012a. Estimation of the energy storage requirement of a future 100% renewable energy system in Japan. Energy Policy 47, 22–31.Esteban, Zhang, Longarte-Galnares, 2012b. Cost-benefit analysis of a green electricity system in Japan considering the indirect economic impacts of tropical cyclones. Energy

Policy 43, 49–57.FEPC. 2011. The Electricity Review in Japan. Federation of Electric Power Company of Japan.FEPC. 2012a.The Electricity Review in Japan. Federation of Electric Power Company of Japan.FEPC.2012b. [Denryoku Toukei Jouhu] [Power Statistics] (in Japanese). Federation of Electric Power Companies of Japan. Available at: ⟨http://www.fepc.or.jp/⟩

[accessed 22 November 2012].FEPC 2013, Nuclear Waste Management. Available at: ⟨http://www.fepc.or.jp/english/nuclear/waste_management/index.html⟩ [accessed April 18, 2013].Goedkoop, et al., 2008. A Life Cycle Impact Assessment Method Which Comprises Harmonized Category Indicators at the Midpoint and the Endpoint Level. PRé Consultants,

Radboud University Nijmegen, Leiden University, RIVM.Gómez, Watterson, 2006. 2006 IPCC Guidelines for National Greenhouse Gas Inventories, vol. 2 Energy, Chapter 2 Stationary Sources. Intergovernmental Panel on Climate

Change, Geneva.Hatamura et al., 2012. Final Report of the Investigation Committee on the Accident at Fukushima Nuclear Power Stations of Tokyo Electric Power Company. National Diet of

Japan.Hondo, 2005. Life cycle GHG emission analysis of power generation systems: Japanese case. Energy 30, 2042–2056.IEA, 2010. Electricity Information. International Energy Agency. Paris.IEEJ, 2011. Japan Energy Brief No. 14. Institute of Energy Economics of Japan.IEEJ, 2012. Energy Indicators of Japan. Available at: ⟨http://eneken.ieej.or.jp/en/⟩ [accessed 22 November 2012].IPSS, 2012. Population Projections for Japan: 2011 to 2060. National Institute of Population and Social Security Research. Tokyo.IPSS, 2013. Population Statistics of Japan. Available at: ⟨http://www.ipss.go.jp/ p-info/e/psj2012/PSJ2012.asp⟩ [accessed 13 April 2013].ISO, 2006.Environmental Management – Life Cycle Assessment – Principles and Framework. ISO 14040:2006. International Organization of Standardization.JCER, 2013. The 39th Medium-Term Economic Forecast (2012 FY-2025FY): Japan's Changes in Industrial Structure and Economic Growth. Tokyo.Kingston, J., 2012. Power politics: Japan's resilient nuclear village. The Asia-Pacific Journal 10 (Issue 43), No. 1.Mainichi Daily, 2012. Japan to Aim for Nuclear Phaseout in 2030s: Draft Energy Policy. Mainichi Daily, 13 September. Available at: ⟨http://mainichi.jp/english/english/

newsselect/news/20120913p2g00m0dm038000c.html⟩ [accessed 14 September 2012].Matsuo, 2012. [kosutotou kenshou iinkai ni yoru gennshiryoku hastuden no kosuto shisan no gaiyou to

hyouka] [Outline and Assessment of Nuclear Power Generation Cost Estimations by the Cost Validation Council] (in Japanese). Institute of Energy Economics of Japan.METI, 2005. Electricity Business Act. Act No. 87 of 2005. Ministry of EconomyTrade and Industry.Government of Japan.METI, 2010. The Strategic Energy Plan of Japan. Ministry of Economy, Trade, and Industry. Government of Japan.Nakata, 2002. Analysis of the impacts of nuclear phase-out on energy systems in Japan. Energy 27, 363–377.Öko Institute, 2011. Global emission model for integrated systems (GEMIS) version 4.7. Germany. Available at: ⟨http://www.oeko.de/service/gemis/en/index.htm⟩ [accessed 10

June 2012].Onoue, Murakami, Sofronis, 2012. Japan's energy supply: mid-to-long-term scenario proposal for a new energy supply system in the aftermath of the March 11 earthquake.

International Journal of Hydrogen Energy 37 (10), 8123–8132.PRC, 2012. Japanese Wary of Nuclear Energy: Disaster ‘Weakened' Nation, Pew Research Center.Schaede, 2012. Japan's Energy Situation and Its Economic Impact—A View from Abroad. In Energy Policy Roundtable 3. Todai Alternative Policy Research Institute. Tokyo,

Japan, 12 July 2012.Short, Packey, Holt, 1995. A Manual for the Economic Evaluation of Energy Efficiency and Renewable Energy Technologies, Report no. NREL/TP-462-5173. National Renewable

Energy Laboratory.Solomon, et al., 2007. Contribution of Working Group I to the Fourth Assessment Report. Intergovernmental Panel on Climate Change. Cambridge University Press, United

Kingdom and New York, NY, USA.Supply and Demand Verification Committee, 2012. [Denryoku cosuto no yokuseisaku ni tsuite] [Measures for the suppression of power costs] (in

Japanese). National Policy Unit. Government of Japan. Available at: ⟨http://www.npu.go.jp/policy/policy09/pdf/20121024/shiryo3.pdf⟩ [accessed 22 November 2012].Takase, Suzuki, 2010. The Japanese energy sector: current situation, and future paths. Energy Policy 39 (11), 6731–6744.Tidball, Bluestein, Rodriguez, Knoke, 2010. Cost and Performance Assumptions for Modeling Electricity Generation Technologies, Report no. NREL/SR-6A20-48595. National

Renewable Energy Laboratory.Valentine, 2011. Emerging symbiosis: renewable energy and energy security. Renewable and Sustainable Energy Reviews 15 (9), 4572–4578.Varun, Bhat, Prakash, 2009. LCA of renewable energy for electricity generation systems. Renewable and Sustainable Energy Reviews 13, 1067–1073.Vivoda, 2012. Japan's energy security predicament post-Fukushima. Energy Policy 46, 135–143.Weisser, 2007. A guide to life-cycle greenhouse gas (GHG) emissions from electric supply technologies. Energy 32, 1543.Williams, 1947. Six-tenths factor aids in approximating costs. Chemical Engineer 54 (124–125), 124–125.World Bank, 2013. World Development Indicators, The World Bank. Available at: ⟨http://data.worldbank.org/data-catalog/world-development-indicators⟩ [accessed 12 April

2013].Zhang, et al., 2012. Scenario analysis on future electricity supply and demand in Japan. Energy 38, 376–385.